# Search the Community

Showing results for tags 'dequation'.

• ### Search By Tags

Type tags separated by commas.

### Forums

• Altair Support Forum
• Welcome to Altair Support Forum
• Installation and Licensing
• Modeling & Visualisation
• Solvers & Optimization
• Multi Body Simulation
• Conceptual design and Industrial design
• Model-Based Development
• Manufacturing Simulation
• CAE Process Automation
• APA - Composites
• APA - CFD & Thermal
• APA - Vehicle Dynamics
• APA - Manufacturing
• APA - Crash and Safety
• APA - Noise, Vibration and Harshness
• APA - System Level Design
• APA - Structural and Fatigue
• APA - Marine
• APA - Optical Design
• Japanユーザーフォーラム
• ユーザーフォーラムへようこそ
• Altair製品の意外な活用例
• インストールとライセンス
• モデリング（プリプロセッシング）
• シミュレーション技術（ソルバー）
• データ可視化（ポストプロセッシング）
• モデルベース開発
• コンセプト設計と工業デザイン
• 製造シミュレーション
• CAE プロセスの自動化
• 学生向け無償版
• エンタープライズソリューション

• 0 Replies

• 0 Reviews

### Organization

Found 4 results

1. Hi, We are trying to solve a optimization problem for random analysis. We want to create a response which can extract MAX stress of all elements selected Responses for stresses is created using DRESP1, Can DERESP2 be used to fetch MAX stress value using dequations. Please help.
2. Hello, this is my first post. I am working on a theoretical problem. Lets say I have some design space for a bracket and I know the geometric locations of the interface connections, like bolt holes or something similar. Optimization Goal: Minimize reaction forces at interfaces due to load Constraints: Volfrac = <1.0 So I created a design variable with an equation for the total reaction force. Its a sum of the reaction forces at each bolt hole and is shown below. To get the bolt hole forces, i did the following. The bolt holes have RBE2 elements connected to springs that are fixed on the opposite end. For each spring there is a total displacement optimization response (the xi values below) for the unfixed end of the springs. I'm using the known stiffness of the springs and the displacement responses to calculate the forces in the springs. (F=K*d) Im currently trying to get this to work for one loadstep, but I was wondering... What if I want to do a series of loadsteps, and weight them like a WCOMP response. Is there a way with the dequation or some other card to do a weighted set of loadcases and minimize the reaction forces like I have described above? If I have orthoganal loads, the optimization solution will be very different for each loadstep. Thanks, Clint
3. Hi, By trying to compute the buckling of stiffened panel, I encountered problems. I have obtained following errors when I have performed a parametric optimization with DRESP1 INERTIA and STRESS in deqatn called by DESSUB. (OptiStruct 12.0.210 ) Here after an extract from the input file which is attached in the totality to this message: DESOBJ(MIN)=24 SUBCASE 7 SPC = 26 LOAD = 25 DESSUB = 2 ... BEGIN BULK ... DRESP2 47 R2_1 1 + DRESP1 27 + DRESP2 46 DRESP2 46 FE2_1 2 + DRESP1 25 DRESP1 27 C2_1 STRESS PBEAML SNMAX 5 DRESP1 25 Ir2_1 INERTIA PBEAML ZZ 5 DRESP1 24 Mass MASS DEQATN 2 F(I)=9.8696*70000*I/530**2/186 DEQATN 1 R(C,F)=F/C DCONSTR 1 47 1.0 DCONADD 2 1 ... Here, the error codes coming from .out *** ERROR # 1803 *** INERTIA responses are not subcase dependent. DRESP1 ID = 25 *** ERROR # 1810 *** Number of responses in DRESP2 inconsistent. DRESP2 ID = 47 This error occurs in module "dppdrv1". I think to have deduced the following fact : *** ERROR # 1803 *** owing to INERTIA card which seems be compatible with DESGLB but in my case, all the subcases are not concerned by this contraints. *** ERROR # 1810 *** owing to STRESS response which seems not be detected by optistruct. Does anyone have a solution or a explanation for this ? Thank MC MEF_7_Colonne.fem
4. Hello, i want to reduce the dynamic stiffness of a rubber bearing by changing the geometry. I have an approximate idea of the "optimized" geometry of this rubber bearing therefore I created morphvolumes and changed the specified area (solid elements) by translating the handles. I saved these changes as shapes which are my design variables. I know that the optimum is between the initial model and the changed model. I want to do an optimization with optistruct to find the optimum. To reduce the dynamic Stiffness i need 3 responses and an equation (dont know another way so far). The calculation of the dynamic Stiffness is simple: Cdyn = Force/Displacement. First i defined an equation with the dequation-panel: F(x,y)=x/y --> where x is the "frf force (ELFORCE)" and y the displacement (frf displacement) which are two of the three responses. The Last response is "function" which includes the dequation and the other two responses defined before. I would like to minimize the response "function" (which is the dynamic stiffness). But it's not possible to create an objective which minimizes my response "function", i get following error: *** ERROR # 1812 *** Multiple responses are assigned to the DESOBJ objective function. A MINMAX/MAXMIN objective function definition with DOBJREF should be used instead. Number of responses = 1000 1) Do you know what the maximum number of responses is? (It's an frequency response analyis from 0 Hz to 3000 Hz with an increment of 3, thats why its 1000 responses (i guess)) . I created an obj reference "dobjref" which includes the response "function", then i created MINMAX objective.This works, and i can start the optimization but here i have some questions: 2) When i define the obj reference i can choose neg and pos reference. What can i do with these references? 3) I don't really know what this minmax function does. Does it minimize the maximum value of the equation? It is not my goal to minimize only the maximum value of the equation, it's rather minimizing the area of the function. 4) Is it normal that optistruct finds a solution after 2 iterations? I have two different shapes and i would expect that it takes a lot of time and iterations to find the optimum. I guess i did something wrong or forgot sth. Sry for asking so many questions and sry for my bad english. I thank you so much in advance. Regards, Fatih Uysal
×
×
• Create New...